This Annual Report on Form 10-K may contain forward-looking statements within the meaning of Section 27A of the Securities Act of 1933, and Section 21E of the Securities Exchange Act of 1934. These statements relate to future events or our future financial performance. We have attempted to identify forward-looking statements by terminology including “anticipates,” “believes,” “can,” “continue,” “could,” “estimates,” “expects,” “intends,” “may,” “plans,” “potential,” “predicts,” “should” or “will” or the negative of these terms or other comparable terminology. These statements are only predictions and involve known and unknown risks, uncertainties and other factors, including the risks outlined under “Factors Affecting Operating Results,” contained“Item 1A. Risk Factors” in Item 7 — “Management’s Discussion and Analysis of Financial Condition and Results of Operation,”this Annual Report, that may cause our actual results, levels of activity, performance or achievements to be materially different from any future results, levels or activity, performance or achievements expressed or implied by these forward-looking statements. Although we believe that the expectations reflected in the forward-looking statements are reasonable, we cannot guarantee future results, levels of activity, performance or achievements. Accordingly, you should not unduly rely on these forward-looking statements, which speak only as of the date of this Annual Report. We are not under any duty to update any of the forward-looking statements after the date we file this Annual Report on Form 10-K or to conform these statements to actual results, unless required by law. You should, however, review the factors and risks we describe in the reports we file from time to time with the Securities and Exchange Commission.

      Illumina®Illumina^®, Array of Arrays^tm, BeadArray^tm, DASL^tm®, GoldenGate®GoldenGate^®, Infinium^tm, Sentrix®Sentrix^® and Oligator®Oligator^® are our trademarks. This report also contains brand names, trademarks or service marks of companies other than Illumina, and these brand names, trademarks and service marks are the property of their respective holders.

      Our annual report on Form 10-K, quarterly reports on Form 10-Q, current reports on Form 8-K, and all amendments to those reports are available free of charge on our website,www.illumina.com. www.illumina.com. The information on our website is not incorporated by reference into this report. Such reports are made available as soon as reasonably practicable after filing with, or furnishing to, the Securities and Exchange Commission. The SEC also maintains an Internet site at www.sec.gov that contains reports, proxy and information statements, and other information regarding issuers that electronically file with the SEC.

      We are a leading developerwere incorporated in April 1998. We develop and marketer ofmarket next-generation tools for the large-scale analysis of genetic variation and function. Understanding genetic variation and function is critical to the development of personalized medicine, a key goal of genomics. Our tools provide information that could be used to improve drugs and therapies, customize diagnoses and treatment, and cure disease.

      In 2001, we began commercial sale of short pieces of DNA called oligonucleotides, which we refer to help achieve the goals of molecular medicine.as oligos, manufactured using our proprietary Oligator technology. We believe our Oligator technology is more cost effective than competing technologies, and this advantage enabled us to market our oligos under a price leadership strategy while still achieving attractive gross margins.

      Our patented BeadArray technology uses microscopic beads randomly deposited in wells to achieve a level of miniaturization that allows for a new scale of experimentation. A microarray is a collection of miniaturized test sites arranged on a surface that permits many tests, or assays, to be performed in parallel. We assemble our arrays using relatively inexpensive materials. Our proprietary manufacturing process allows us to easily adapt the arrays to a broad range of applications, including both genotyping and gene expression. These characteristics allow us to create next-generation arrays with a unique combination of high throughput, cost effectiveness and flexibility. In addition, our complementary Oligator technology permits parallel synthesis of the millions of different pieces of DNA necessary to perform large-scale genetic analysis on arrays.

      Our production-scale BeadLab is a turnkey platform that includes all hardware and software necessary to enable researchers to perform genetic analysis research on what we believe is an internationally funded successor project to the Human Genome Project that is creating a map of genetic variations that may be used to perform disease-related research.unprecedented scale. This map of the human genome will allow more rapid and efficient large-scale genetic association studies aimed at discovering variants contributing to human disease and differential response to drug treatments. We are one of five funded first phase U.S. participants in a worldwide initiative that includes research groups in Canada, China, Japan, Nigeria and the United Kingdom. We are directly responsible for screening over 15% of the assays in the first phase of this project. This effort leverages our Oligator DNA synthesis capability and the production-scale throughput of our genotyping services operation. Our BeadLabsystem is being used by organizations responsible for creating over 60%marketed to a small number of the assays in the first phasehigh-throughput genotyping users. As of this project, whichJanuary 1, 2006, we expect to be completed in early 2005.

      In 2003, we announced the launch of several new products, including 1) a new array format, the Sentrix BeadChip, which significantly expands market opportunities for our BeadArray technology and provides increased experimental flexibility for life science researchers; 2) a gene expression product line on both the Sentrix Array Matrix and the Sentrix BeadChip that allows researchers to analyze a focused set of genes across eight to 96 samples on a single array; and 3) a benchtop SNP genotyping and gene expression system, the BeadStation, for performing moderate-scale genotyping and gene expression using our technology. The BeadStation includes our BeadArray Reader, analysis software and assay reagents and is designed to match the throughput requirements and variable automation needs of individual research groups and core labs. Sales of these products began in earlythe first quarter of 2004 and, as of January 2, 2005,1, 2006, we have shipped 42115 BeadStations.

      In late 2004, we announced a strategic collaboration with Invitrogen Corporation (“Invitrogen”) to synthesize and distribute oligos. In the launchthird quarter of new2005, we began shipping oligo products in connection with this agreement. As part of the agreement, we have developed the next generation of our Oligator DNA synthesis technology, which we have designed to support both plate- and tube-based capabilities. Invitrogen is responsible for sales, marketing and technical support. Profits from sales of collaboration products are divided equally between the two companies.

      In 2005, we began shipments of Sentrix BeadChips for whole-genome gene expression and whole-genome genotyping. The whole-genome gene expression BeadChips are designed to enable high-performance, cost-effective, whole-genome expression profiling of multiple samples on a single chip, resulting in a dramatic reduction in cost of whole-genome expression analysis while allowing researchers to expandanalysis. Our whole-genome expression product line includes multi-sample products for both the scaleHuman and reproducibility of large-scale biological experimentation.Mouse Genomes. The whole-genome genotyping BeadChip can be scaledis designed to unlimitedscale to high levels of multiplexing without compromising data quality and willto provide scientists the ability to query hundreds of thousands of SNPs in parallel, a high-value setparallel. In the second quarter of over 100,000 SNPs. In 2004,2005, we also announced two new versions of the Sentrix Array Matrix designed for researchers who want to take advantagecommenced shipment of our technology, but whose projects require fewerfirst whole-genome genotyping BeadChip, the HumanHap1, which interrogates more than 100,000 SNPs per sample than the number utilized on our standard 1536-plex array products.in parallel.

      In late 2004, we announced a strategic collaboration with Invitrogen Corporation to synthesize and distribute oligos. Under the agreement, we intend to expand our Oligator DNA synthesis technology, and Invitrogen will be responsible for sales, marketing and technical support. Profits from sales of collaboration products will be divided equally between the two companies.

      We are seeking to continue to expand our customer base for our BeadArray technology; however, we can give no assurance that our sales efforts will continue to be successful.

      We were incorporated in California in April 1998. We reincorporated in Delaware in July 2000. Our principal executive offices are located at 9885 Towne Centre Drive, San Diego, California 92121. Our telephone number is (858) 202-4500.

      Every person inherits two copies of each gene, one from each parent. The two copies of each gene may be identical, or they may be different. These differences are referred to as genetic variation. Examples of the physical consequences of genetic variation include differences in eye and hair color. Genetic variation can also have important medical consequences, including predisposition to disease and differential response to drugs. Genetic variation affects diseases,disease susceptibility, including predisposition to cancer, diabetes, cardiovascular disease and Alzheimer’s disease. In addition, genetic variation may cause people to respond differently to the same drug.drug treatment. Some people may respond well, others may not respond at all, and still others may experience adverse side effects. The mostA common form of genetic variation is a Single Nucleotide Polymorphism, or SNP. A SNP is a variation in a single position in a DNA sequence. It is estimated that the human genome contains between three and sixover nine million SNPs.

      While in some cases a single SNP will be responsible for medically important effects, it is now believed that combinations of SNPs may contribute to the genetic componentdevelopment of most major diseases is the resultdiseases. Since there are millions of the interaction of many SNPs. Therefore,SNPs, it is important to investigate many representative, well-chosen SNPs togethersimultaneously in order to discover medically valuable information.

      Current effortsAnother contributor to understand genetic variationdisease and function have primarily centered around SNP genotypingdysfunction is the over- or under-expression of genes within an organism’s cells. A very complex network of genes interacts to produce healthy individuals. The challenge for scientists is to delineate the associated genes’ expression patterns and gene expression profiling.their relationship to disease. Until recently, this problem was addressed by investigating effects on a gene-by-gene basis. This is time consuming, and difficulties exist when several pathways can not be observed or “controlled” at the same time. With the advent of microarray technology, thousands of genes can now be tested at the same time.

      SNP genotyping is the process of determining which SNPs arebase (A, C, G or T) is present at a particular site in each of the two copies of a gene, or other portion of DNA sequence,genome within an individual or other organism. The use of SNP genotyping to obtain meaningful statistics on the effect of an individual SNP or a collection of SNPs, and to apply that information to clinical trials and diagnostic testing, requires the analysis of millions of SNP genotypes and the testing of large populations for each disease. For example, a single large clinical trial could involve genotyping 200,000300,000 SNPs per patient in 1,000 patients, thus requiring 200300 million assays. Using previously available technologies, this scale of SNP genotyping was both impractical and prohibitively expensive.

      Large-scale SNP genotyping willcan be used forin a variety of applications,ways, including studies designed to understand the genetic contributions to disease (disease association studies), genomics-based drug development, clinical trial analysis, disease predisposition testing, and disease diagnosis. SNP genotyping can also be used outside of healthcare, for example in the development of plants and animals with desirable commercial characteristics. These markets will require billions of SNP genotyping assays annually.

      Gene expression profiling is the process of determining which genes are active in a specific cell or group of cells and is accomplished by measuring mRNA, the intermediary messenger between genes (DNA) and proteins. Variation in gene expression can cause disease, or act as an important indicator of disease or predisposition to disease. By comparing gene expression patterns between cells from different environments, such as normal tissue compared to diseased tissue or in the presence or absence of a drug, specific genes or groups of genes that play a role in these processes can be identified. Studies of this type, often used in drug discovery, require monitoring thousands, and preferably tens of thousands, of mRNAs in large numbers of samples. Once a smaller set of genes of interest has been identified, researchers can then examine how these genes are expressed or suppressed across numerous samples, for example, within a clinical trial. The high cost of current gene expression methods has limited the development of the gene expression market.

      As gene expression patterns are correlated to specific diseases, gene expression profiling is becoming an increasingly important diagnostic tool. Diagnostic use of expression profiling tools is anticipated to grow rapidly with the combination of the sequencing of various genomes and the availability of more cost-effective technologies.

      We have developed a proprietary array technology that enables the large-scale analysis of genetic variation and function. Our BeadArray technology combines microscopic beads and a substrate in a simple proprietary manufacturing process to produce arrays that can perform many assays simultaneously. Our BeadArray technology provides a unique combination of high throughput, cost effectiveness, and flexibility. We achieve high throughput with a high density of test sites per array and our abilitywe are able to format arrays either in either a pattern arranged to match the wells of standard microtiter plates or in various configurations in the format of standard microscope slides. We seek to maximize cost effectiveness by reducing consumption of expensive reagents and valuable samples, and fromthrough the low manufacturing costs associated with our technologies. Our ability to vary the size, shape and format of the well patterns and to create specific bead pools, or sensors, for different applications provides the flexibility to address multiple markets and market segments. We believe that these features have enabled our BeadArray technology to become a leading platform for the emerging high-growth market of SNP genotyping and expect they will enable us to become a key player in the gene expression market.

      Our proprietary BeadArray technology combines microwells etched into a substrate and specially prepared beads that self-assemble into an array. We have deployed our BeadArray technology in two different Sentrix array formats, the Array Matrix and the BeadChip. Our first bead-based product was the Array Matrix which incorporates fiber optic bundles. We have theThe fiber optic bundles, manufactured to our specifications, which we cut into lengths of less than one inch.inch, are manufactured to our specifications. Each bundle containsis comprised of approximately 50,000 individual fibers and 96 of these bundles are placed into an aluminum plate, which forms an Array Matrix. BeadChips are fabricated in microscope slide-shaped sizes with varying numbers of sample sites per slide. Both formats are chemically etched to create tens to hundreds of thousands of wells for each sample site.

      In a separate process, we create sensors by affixing a specific type of molecule to each of the billions of microscopic beads in a batch. We make different batches of beads, with the beads in a given batch coated with one particular type of molecule. The particular molecules on a bead define that bead’s function as a sensor. For example, we create a batch of SNP sensors by attaching a particular DNA sequence, or oligo, to each bead in the batch. We combine batches of coated beads to form a pool specific to the type of array we intend to create. A bead pool one milliliter in volume contains sufficient beads to produce thousands of arrays. One of the advantages of this technology is that it allows us to create universal arrays for SNP genotyping, and by varying the reagent kit, still bewe are able to use the array to test for any combination of SNPs.

      To form an array, a pool of coated beads is brought into contact with the array surface where they are randomly drawn into the wells, one bead per well. The tens of thousands of beads in the wells comprise our individual arrays. Because the beads assemble randomly into the wells, we perform a final procedure called decoding’decoding’ in order to determine which bead type occupies which well in the array. We employ several proprietary methods for decoding, a process that requires only a few steps to identify all the beads in the array. One beneficial by-product of the decoding process is a validation of each bead in the array. This quality control test characterizes the performance of each bead and can identify and eliminate use of any empty wells. We ensure that each bead type on the array is sufficiently represented by havingincluding multiple copies of each bead type. This improvesMultiple bead type copies improve the reliability and accuracy of the resulting data by allowing statistical processing of the results of identical beads. We believe we are the only microarray company to provide this level of quality control in the industry.

      An experiment is performed on the Array Matrix by preparing a sample, such as DNA from a patient, and introducing it to the array. The design features of our Array Matrix allow it to be simply dipped into a solution containing the sample, whereas our BeadChip allows processing of samples on a slide.slide-sized platform. The molecules in the sample bind to their matching molecules on the coated bead. The BeadArray Reader detects the matched molecules by shining a laser on the fiber optic bundle or on the BeadChip. Since the molecules in the sample have a structure that causes them to emit light in response to a laser, detection of a binding event is possible. This allows the measurement of the number of molecules bound to each coated bead, resulting in a quantitative analysis of the sample.

      Genomic applications require many different short pieces of DNA that can be made synthetically, called oligonucleotides.oligos. For example, SNP genotyping typically requiresmay require three to four different oligonucleotidesoligos per assay. A SNP genotyping experiment analyzing 10,000 SNPs may therefore require 30,000 to 40,000 different oligonucleotides,oligos, contributing significantly to the expense of the experiment.

      We have designeddeveloped our proprietary Oligator technology for the parallel synthesis of many different oligonucleotidesoligos to meet the requirements of large-scale genomics applications. We believe that our Oligator technology is substantially more cost effective and provides significantly higher throughput than available commercial alternatives. Our synthesis machines are computer controlled and utilize many robotic processes to minimize the amount of labor used in the manufacturing process. Each of these synthesizers can produce up to 3072 oligos in parallel, using very small amounts of material. We believe both of these attributes are substantial improvements over other existing technologies. In 2005, we intend to implementimplemented our fourth-generation Oligator technology, which will further expandis capable of manufacturing up to 13,000 different oligos per run. This is an improvement over prior generations of technology where we could only manufacture approximately 3,000 oligos per run. This increase in scale was necessary to enable us to support the manufacture of oligos under our production capabilities and extendcollaboration with Invitrogen as well as to support our technology into tube-based oligo products.increased need for oligos, a critical component of our BeadArray technology.

      We believe that our BeadArray and Oligator technologies provide distinct advantages, in a variety of applications, over competing technologies, by creating cost-effective, highly miniaturized arrays with the following advantages:

     High Throughput. The miniaturization of our BeadArray technology provides significantly greatervery high information content per unit area than any other array known to us.area. To further increase sample throughput, we have formatted our arraysarray matrix in a pattern arranged to match the wells of standard microtiter plates, allowing throughput levels of up to nearly 150,000 unique assays per microtiter plate, as well as theand we use of laboratory robotics to speed process time. Similarly, we have patterned our whole-genome expression BeadChips to support up to 48,000 gene expression assays for six samples with each BeadChip. The Oligator’s parallel synthesis capability allows us to manufacture the diversity of oligonucleotidesoligos necessary to support large-scale genomic applications.

     Cost Effectiveness. Our array products substantially reduce the cost of experiments as a result of our proprietary manufacturing process and our ability to capitalize on cost reductions generated by advances in fiber optics, plasma etching processes, digital imaging and bead chemistry. In addition, these products require smaller reagent volumes than other array technologies, and therefore reduce reagent costs. Our cost-effective Oligator technology further reduces reagent costs, as well as the cost of coating beads.

     Flexibility. A wide variety of conventional chemistries are available for attaching different molecules, such as DNA, RNA, proteins, and other chemicals to beads. By using beads, we are able to take advantage of these chemistries to create a wide variety of sensors, which we assemble into arrays using the same proprietary manufacturing process. In addition, we can have fiber optic bundles and BeadChips manufactured in multiple shapes and sizes with wells organized in various arrangements to optimize them for different markets and market segments. In combination, the use of beads and etched wells provides the flexibility and scalability for our BeadArray technology to be tailored to perform many applications in many different market segments, from drug discovery to diagnostics. Our Oligator technology allows us to manufacture a wide diversity of lengths and quantities of oligonucleotides.oligos.

     Quality. The quality of our products is dependent upon each element in the system, the array, the assay used to perform the experiment and the instrumentation and software used to capture the results.

      Each array is manufactured with a high density of beads in each arraywhich enables us to have multiple copies of each individual bead type. We measure the copies simultaneously and combine them into one data point. This allows us to make a comparison of each bead against its own population of identical beads, which permits the statistical calculation of a more reliable and accurate value for each data point. Finally, the manufacture of the array includes a proprietary decoding step that also functions as a quality control test of every bead on every array, improving the overall quality of the data.

      Our goal is to make our BeadArray platforms the industry standard for products and services utilizing array technologies. We plan to achieve this by:

      The first implementation of our BeadArray technology, the Sentrix Array Matrix, is a disposable matrix with 96 fiber optic bundles arranged in a pattern that matches the standard96-well microtiter plate. Each fiber optic bundle performs more than 1,500 unique assays. Therefore, one Sentrix Array Matrix can perform nearly 150,000 individual assays simultaneously, more than any other commercial array system known to us. The BeadChip, introduced in 2003, is fabricated in multiple configurations to support multiple applications and scanning technologies.

      We have provided genotyping services using our proprietary BeadArray technology since 2001. In addition, we have developed our first genotyping and gene expression products based on this technology. These products include disposable Sentrix Array Matrices and BeadChips, GoldenGate and Infinium reagent kits for SNP genotyping, BeadArray Reader scanning instruments and an evolving portfolio of custom and standard gene expression products.

      In 2001, we introduced the first commercial application of our BeadArray technology by launching our SNP genotyping services product line. Since this launch, we have had peak days in which we operated at over two million genotypes per day based on individual samples. To our knowledge, no other genotyping platform can achieve comparable levels of throughput while delivering such high accuracy and low cost.

      We designed our first consumable BeadArray product, the Sentrix Array Matrix, for SNP genotyping. The Sentrix Array Matrix uses a universal format that allows it to analyze any set of SNPs. We have also developed reagent kits based on GoldenGate assay protocols and the BeadArray Reader, a laser scanner, which is used to read our array products.

      Depending on throughput and automation requirements, our customers can select the system configuration to best meet their needs. For production-scale throughput, our BeadLab would be appropriate, and for moderate-scale throughput, our BeadStation would be selected. Our BeadLab includes our BeadArray Reader, combined with LIMS, standard operating procedures and analytical software and fluid handling robotics. This production-scale system was commercialized in late 2002 and when installed, this system can routinely produce up to 1.4 millionmillions of genotypes per day.

      The BeadStation, a system for performing moderate-scale genotyping designed to match the throughput requirements of individual research groups and core labs, was commercialized in late 2003. The BeadStation includes our BeadArray Reader and genotyping and/or gene expression analysis software. Our BeadStations are fully upgradeable to a full BeadLab through various steps that add automation, sample preparation equipment and LIMS capability.

      Also inIn 2003, we announced the availability of an assay set for genetic linkage analysis. This standard product has been deployed in our genotyping services operation and is also sold to customers who use our SNP genotyping systems. Genetic linkage analysis can help identify chromosomal regions with potential disease associations across a related set of samples.

      In 2004, we announced a new Sentrix Human-1 Genotyping BeadChip for whole-genome genotyping. This BeadChip will provideprovides to scientists the ability to interrogate over 100,000 SNPs located in high-value genetic regions of the human genome. In 2006, we announced the Sentrix HumanHap300 Genotyping BeadChip for larger-scale SNPs studies, which can assay more than 317,000 SNPs displayed across the entire human genome.

      With the addition of application specific accessory kits, our production-scale BeadLabs and BeadStations are capable of performing a growing number of applications, including gene expression profiling.

      In 2003, we introduced our focused set gene expression products on both the Sentrix Array Matrix and Sentrix BeadChip platforms. For high-throughput projects, ourOur system includes a BeadArray Reader for imaging Sentrix Array Matrices and BeadChips, a hybridization chamber and software for data extraction. For research projects that require moderate throughput, a version of the Sentrix BeadChip analyzes eight samples in parallel and can be scanned on a portion of the installed base of Axon Instruments’ GenePix^tm scanners. In addition, we have developed standard gene expression products for each of the human, mouse and arabidopsis genomes.genomes with an additional panel that focuses on human toxicology.

      In 2004,2005, we announcedbegan shipment of the Sentrix Human-6 and HumanRef-8 Expression BeadChip products. Both products will allow large-scale expression profiling of multiple samples on a single chip and are imaged using our BeadArray Reader. The Human-6 BeadChip is designed to analyze six discrete whole-human-genome samples on one chip, interrogating in each sample approximately 48,000 transcripts from the estimated 30,000 genes in the human genome. The HumanRef-8 BeadChip product analyzes eight samples in parallel against 24,000 transcripts from the roughly 22,000 genes represented in the consensus RefSeq database, a well-characterized whole-genome subset used broadly in genetic analysis. We expect that these new gene expression BeadChips will dramatically reduce the cost of whole-genome expression analysis, allowing researchers to expand the scale and reproducibility of large-scale biological experimentation.

      The BeadArray Reader, an instrument we developed, is a key component of both our production-scale BeadLab and our benchtop BeadStation. This scanning equipment uses a laser to read the results of experiments that are captured on our arrays and was designed to be used in all areas of genetic analysis that use our Sentrix Array Matrices and Sentrix BeadChips.

      We have put in place an oligonucleotidea state of the art oligo manufacturing facility. This facility that currently hasserves both the capability of producing approximately 20 million oligonucleotides per year.commercial needs under our collaboration with Invitrogen and our internal needs. In addition to their use to coat beads, these oligonucleotidesoligos are components of the reagent kits for our BeadArray products and are used for assay development. Because our production capacity exceeds our internal needs, we began to offer oligonucleotides for sale to high volume users in 2001. We provide oligonucleotidesmanufacture oligos in a wide range of lengths and in several scales, with the ability to add many types of modifications. We offer a range of quality control options and have implemented a laboratory information management system to control much of the manufacturing process. In 2003, we introduced the first standard product offerings in our Oligator product line, a whole-genome oligonucleotideoligo reference set designed and optimized for spotted gene expression microarrays, and in 2004, we introduced a mouse genome oligo set, also for use on spotted gene expression arrays. We believe our Oligator technology is more cost effective than competing technologies, which has allowed us to market our oligonucleotides under a price leadership strategy while still achieving attractive gross margins. In 2005, in connectionwe stopped selling oligos directly into the market and began shipping oligos under our collaboration with a collaboration agreement with Invitrogen Corporation, we intend to implement fourth-generation Oligator technology which will further expand our production capabilities and extend our technology into tube-based oligo products.Invitrogen.

      In December 2004, we entered into a strategic collaboration with Invitrogen Corporation. Through thisInvitrogen. The goal of the collaboration we intendis to expandcombine our Oligator DNA synthesis technology and combine that capabilityexpertise in oligo manufacturing with Invitrogen’sthe sales, marketing and distribution channels.capabilities of Invitrogen. In connection with the collaboration, we have developed the next generation Oligator^® DNA synthesis technology. This technology includes both plate- and tube-based capabilities. Under the terms of the agreement, Invitrogen paid us an upfront non-refundable collaboration payment of $2.3 million in the first quarter of 2005. Additionally, upon the achievement of a certain milestone, Invitrogen was obligated to make a milestone payment of $1.1 million to us. As of January 1, 2006, this milestone has agreed to pay us up to $3.4 million, which we planbeen achieved and the milestone payment was received. We have used these funds to invest in our San Diego facility to enable the development and implementation of fourth-generation Oligator technology and to extend the technology into tube-based oligo products. We began manufacturing and shipping the plate-based and certain tube-based oligo products under the collaboration in the third quarter of 2005. In addition, the agreement provides for the transfer of our Oligator technology into two Invitrogen facilities outside North America. ProfitCollaboration profit from the sale of collaboration products will be divided equally between the two companies.

      We have made substantial investments in research and development since our inception. We have assembled a team of skilled engineers and scientists who are specialists in biology, chemistry, informatics, instrumentation, optical systems, software, manufacturing and other related areas required to complete the development of our products. Our research and development efforts have focused primarily on the tasks required to optimize our BeadArray and Oligator technologies and to support commercialization of the products and services derived from these technologies. These efforts include the following, among others:

      Our research and development expenses for the fiscal years2005, 2004 and 2003 and 2002 (exclusive(inclusive of charges relating to stock basedstock-based compensation of $0.8$0.1 million, $1.3$0.3 million, and $2.4$1.3 million, respectively) were $27.7 million, $21.1 million and $22.5 million, and $26.8 million, respectively. WeAs compared to 2005, we expect research and development expense to increase in 2005 as compared to 2004absolute dollars during 2006, as we continue to expand our research and product development efforts.efforts, but decrease as a percentage of overall revenue in 2006.

      Government grants allow us to fund internal scientific programs and exploratory research. We retain ownership of all intellectual property and commercial rights generated during these projects, subject to a non-exclusive, non-transferable,paid-up license to practice, for or on behalf of the United States, inventions made with federal funds. This license is retained by the U.S. government as provided by applicable statutes and regulations. We do not believe that the retained license will have any impact on our ability to market our products, and we do not need government approval with respect to this license in order to enter into collaborations or other relationships with third parties. We arewere the recipient of a grant from the National Institutes of Health covering our participation in the first phase of the International HapMap Project, which is a $100.0$100 million, internationally funded successor project to the Human Genome Project that will help identify a map of genetic variations that may be used to perform disease-related research. We could receive up toreceived $9.1 million of funding for this project which coverscovered basic research activities, the development of SNP assays and the genotyping to be performed on those assays. Asassays, all of which was earned in prior years, except for approximately $0.8 million, which was recognized as revenue during the endfirst quarter of 2004, we had approximately $0.7 million of funding remaining related to this project, which is expected to be received in earlyfiscal 2005.

      We have an extensive patent portfolio, including, as of February 1, 2006, ownership of, or exclusive licenses to, 2938 issued U.S. patents and 76102 pending U.S. patent applications, including sevensix allowed applications that have not yet issued as patents, some of which derive from a common parent application. Our issued patents, which cover various aspects of our BeadArray, oligonucleotidearray, assay, oligo synthesis, instrument and chemical detection technologies, expire between 2011 and 2020.2022. We are seeking to extend this patent protection on our BeadArray, DASL, GoldenGate, Infinium, CyVera, Oligator, Sentrix, Array of Arrays and related technologies. We have received or filed counterparts for many of these patents and applications in one or more foreign countries.

      We also rely upon trade secrets, know-how, copyright and trademark protection, as well as continuing technological innovation and licensing opportunities to develop and maintain our competitive position. Our success will depend in part on our ability to obtain patent protection for our products and processes, to preserve our copyrights and trade secrets, to operate without infringing the proprietary rights of third parties and to acquire licenses related to enabling technology or products used with our BeadArray, DASL, GoldenGate, Infinium, Sentrix, Array of Arrays, CyVera and Oligator technologies.

      We are party to various exclusive and non-exclusive license agreements with third parties, which grant us rights to use key aspects of our array technology, assay methods, chemical detection methods, reagent kits and scanning equipment. We have exclusive licenses from Tufts University to patents that cover our use of BeadArray technology. These patents were filed by Dr. David Walt, a member of our board of directors, the Chairman of our Scientific Advisory Board and one of our founders. Our exclusive licenses expire with the termination of the underlying patents, which will occur between 2010 and 2020.2019. In 2001, we entered into a non-exclusive license agreement with Amersham Biosciences that covers certain technology contained in our BeadArray Reader. In 2002, we obtained a non-exclusive license from Dade Behring Marburg GmbH that relates to certain components of our GoldenGate assay. We also have additional nonexclusive licenses from various third parties for other components of our products. In all cases, the agreements remain in effect over the term of the underlying patents, may be terminated at our request without further obligation and require that we pay customary royalties while the agreement is in effect.

      Our current products address the genetic analysis portion of the life sciences market, in particular, experiments involving SNP genotyping and gene expression profiling. These experiments may be involved in many areas of biologic research, including basic human disease research, pharmaceutical drug discovery and development, pharmacogenomics, toxicogenomics and agricultural research. Our potential customers include pharmaceutical, biotechnology, agrichemical, diagnostics and consumer products companies, as well as academic or private research centers. The genetic analysis market is relatively new and emerging and its size and speed of development will be ultimately driven by, among other items:

      We market and distribute our products directly to customers in North America, major European markets, Japan and Singapore. In each of these areas, we have dedicated sales, service and application support personnel responsible for expanding and managing their respective customer bases. In markets outside of these areas, primarily the Pacific Rim countries and Europe, we sell our products and provide services to customers through distributors that specialize in life science products. We expect to significantly increase our sales and distribution resources during 20052006 and beyond as we launch a number of new products and expand the number of customers that can use our products.

      In late 2004, we entered into a strategic collaboration with Invitrogen Corporation with a goal of leveraging our strength in oligo synthesis with Invitrogen’s extensive sales, marketing and distribution channels. We expect to transitiontransitioned all responsibility for oligo sales, marketing and technical support to Invitrogen in the beginning of the third quarter of 2005.

      We manufacture our array platforms, reagent kits, scanning equipment and oligonucleotidesoligos in-house and believe that we currently have the ability to manufacture these in sufficient quantity to meet anticipated internal and external needs.our business plan for 2006. We currently depend upon outside suppliers for materials used in the manufacture of our products. We intend to continue, and may extend, the outsourcing of portions of our manufacturing process to subcontractors where we determine it is in our best commercial interests.

      During 2001, we moved into a new facility which allowed us to design the manufacturing areas to fit our specific processes, and optimize material flow and personnel movement. In addition, we have implemented information management systems for many of our manufacturing and services operations to manage all aspects of material and sample use. We adhere to access and safety standards required by federal, state and local health ordinances, such as standards for the use, handling and disposal of hazardous substances.

      Although we expect that our BeadArray products and services will provide significant advantages over currently available products and services, we expect to encounter intense competition from other companies that offer products and services for the SNP genotyping and gene expression markets. These include companies such as Aclara Biosciences (recently acquired(acquired by ViroLogic), Affymetrix, Agilent, Amersham Biosciences (recently acquired(acquired by GE Corp.) and now named GE Healthcare), Applied Biosystems, Beckman Coulter, Caliper Technologies, Luminex, ParAllele Bioscience (acquired by Affymetrix), Perlegen Sciences, NimbleGen, Sequenom and Third Wave Technologies. ManySome of these companies have or will have substantially greater financial, technical, research, and other resources and larger, more established marketing, sales, distribution and service organizations than we do. In addition, they may have greater name recognition than we do in the markets we need to address and in some cases a large installed base of systems. Each of these markets is very competitive and we expect new competitors to emerge and the intensity of competition to increase in the future. In order to effectively compete with these companies, we will need to demonstrate that our products have superior throughput, cost and accuracy advantages over the existing products. Rapid technological development may result in our products or technologies becoming obsolete. Products offered by us could be made obsolete either by less expensive or more effective products based on similar or other technologies. Although we believe that our technology and products will offer advantages that will enable us to compete effectively with these companies, we cannot assure you that we will be successful.

      We operate in one business segment, for the development, manufacture and commercialization of tools for genetic analysis. Our operations are treated as one segment as we only report operating results on an aggregate basis to chief operating decision makers of Illumina.

      During 2004, $26.42005, $28.0 million, or 52%38%, of our total revenuesrevenue came from customers outside the United States, as compared to $14.4$26.4 million, or 51%52%, in 2003.2004. Sales to territories outside of the United States are generally denominated in U.S. dollars. We expect that sales to international customers will continue to be an important and growing source of revenues.revenue. We have sales support resources in Western Europe and direct sales offices in Japan, Singapore and Singapore.China. In addition, we have distributor relationships in various countries in the Pacific Rim region.region and Europe.

      Information about the geographies in which we operate can be found in the Notesnotes to Consolidated Financial Statementsthe consolidated financial statements at Note 11, “Segment Information, Geographic Data and Significant Customers.”

      Historically, customer purchasing patterns have not shown significant seasonal variation, although demand for our products is usually lowest in the first quarter of the calendar year and highest in the fourth quarter of the calendar year as academic customers spend unused budget allocations before the end of the government’s fiscal year.

      We are dedicated to the protection of our employees and the environment. Our operations require the use of hazardous materials which subject us to a variety of federal, state and local environmental and safety laws and regulations. We believe we are in material compliance with current applicable laws and regulations; however, we could be held liable for damages and fines should contamination of the environment or individual exposures to hazardous substances occur. In addition, we cannot predict how changes in these laws and regulations, or the development of new laws and regulations, will affect our business operations or the cost of compliance.

      As of January 2, 2005,1, 2006, we had a total of 278375 employees, 6073 of whom hold Ph.D. degrees and 37degrees. 44 of suchour employees with Ph.D. degreed employeesdegrees are engaged in full-time research and development activities. None of our employees isare represented by a labor union. We consider our employee relations to be positive.

      Our executive officers as of February 28, 2005,1, 2006, are as follows:

     Jay T. Flatleyhas served as our President, Chief Executive Officer and a Director since October 1999. Prior to joining Illumina, Mr. Flatley was co-founder, President, Chief Executive Officer and a Director of Molecular Dynamics, a life sciences company, from May 1994 to September 1999. He served in various other positions with that company from 1987 to 1994. From 1985 to 1987, Mr. Flatley was Vice President of Engineering and Vice President of Strategic Planning at Plexus Computers, a UNIX computer company. Mr. Flatley also serves as a director at GenVault. Mr. Flatley holds a B.A. in Economics from Claremont McKenna College and a B.S. and M.S. in Industrial Engineering from Stanford University.

     David L. Barker, Ph.D.,Christian O. Henryhas servedjoined Illumina in June 2005 as our Vice President and Chief ScientificFinancial Officer. He is responsible for worldwide financial operations, controllership functions and facilities management. Mr. Henry served previously as the Chief Financial Officer since March 2000.for Tickets.com, a publicly traded, online ticket provider that was recently acquired by Major League Baseball Advanced Media, LP. Prior to joining us, Dr. Barkerthat, Mr. Henry was Vice President, Finance and Chief Science Advisor at Amersham Pharmacia Biotech,Corporate Controller of Affymetrix, Inc., a publicly traded life sciences company, from September 1998 to March 2000. From May 1997 to September 1998, Dr. Barker was Vice President of Researchcompany. He previously held a similar position at Nektar Therapeutics (formerly Inhale Therapeutic Systems, Inc.). Mr. Henry received a BA in biochemistry and Business Development at Molecular Dynamics. From 1992 to 1997, he was Vice President of Scientific Development. From 1988 to 1995, he held various other positions with that company. Dr. Barker holds a B.S. in Chemistry from California Institute of Technology and received his Ph.D. in Biochemistry from Brandeis University.

     Arnold Oliphant, Ph.D.,has served as our Vice President of Scientific Operations since October 2000. Prior to joining us, Dr. Oliphant was Vice President of Functional Genomics at Myriad Genetics, a genomics company, from 1997 to September 2000 and was Process Development and Production Director from January 1995 to June 1997. From January 1992 to January 1995, Dr. Oliphant held several positions at Pioneer Hybrid International, a plant genetics company and prior to that was an Assistant Professor at the University of Utah. Dr. Oliphant received his B.A. in biology from the University of Utah and his Ph.D. in Genetics from the Harvard Medical School.

     John R. Stuelpnagel, D.V.M.,one of our founders, is our Senior Vice President and Chief Operating Officer and has been a director since April 1998. From October 1999 to April 2002, he served as our Vice President of Business Development. From April 1998 to October 1999, he served as our acting President and Chief Executive Officer and was acting Chief Financial Officer through April 2000. While founding Illumina, Dr. Stuelpnagel was an associate with CW Group, a venture capital firm, from June 1997 to September 1998 and with Catalyst Partners, a venture capital firm, from August 1996 to June 1997. Dr. Stuelpnagel received his B.S. in Biochemistry and his Doctorate in Veterinary Medicine from the University of California, Davis and his M.B.A. from the University of California, Los Angeles.

      Our principal research and development, manufacturing and administrative facilities occupy approximately 90,000116,000 square feet of three buildings located in San Diego, California, which we purchased, along with eight acres of adjacent land, in January 2002. In connection with this purchase we assumed a $26$26.0 million,10-year mortgage on the property at a fixed interest rate of 8.36%. In June 2004, we entered into a conditional agreement to sell our land and buildings for $42.0 million and to lease back such property for an initial term of ten years. The sale was completed in August 2004, at which time the lease was signed. Under the terms of the lease, we made a $1.9 million security deposit and are obligated to pay monthly rent of approximately $318,643 for the first year with an annual increase of 3% in each subsequent year. The current monthly rent under this lease is $328,202. The lease contains an option to review for three additional periods of five years each. In January 2006, we began leasing approximately 4,500 square feet of industrial space in San Diego, California to be used for distribution and storage of our products. The initial term of this lease is three years. In conjunction with our acquisition of CyVera in April 2005, we also lease office space for a facility located in Connecticut that occupies 14,884 square feet of office space. This lease is non-cancelable and expires as of April 2008. This facility is used primarily for research and development purposes. We expect that these facilities will be sufficient for our San Diego basedU.S.-based operations for the foreseeable future.through at least 2006.

      In February 2003, the Companywe began leasing approximately 3,300 square feet of office space in Tokyo and, in January 2004, we began leasing approximately 1,600 square feet of office space in Singapore. In November 2005, we began leasing approximately 200 square feet of office space in China. These facilities are used by local sales, marketing and field service personnel.

      InOn July 26, 2004, Affymetrix, Inc. (“Affymetrix”) filed a complaint in the U.S. District Court for the District of Delaware alleging that certainthe use, manufacture and sale of our BeadArray products and services, including the Array Matrix and BeadChip products, infringe six Affymetrix patents. The suitAffymetrix seeks an unspecified amount of monetary damages and a judgment enjoininginjunction against the sale of products, if any, that are determined to be infringing these patents. Inpatents, unspecified monetary damages, interest and attorneys’ fees. On September 15, 2004, we filed our answer and counterclaims to Affymetrix’ complaint, seeking declaratory judgments from the court that we do not infringe the Affymetrix patents, and that such patents are invalid, and filed counterclaims against Affymetrix for unfair competition and interference with actual and prospective economic advantage. On January 7, 2006, we sought leave to file our first amended answer and counterclaims, adding allegations of inequitable conduct with respect to all six asserted Affymetrix patents, violation of Section 2 of the Sherman Act, and unclean hands. Trial is scheduled for October 16, 2006. We believe we have meritorious defenses against each of the infringement claims alleged by Affymetrix and intend to vigorously defend ourselves against this suit. However, we cannot be sure that we will prevail in this matter. Any unfavorable determination, and in particular, any significant cash amounts required to be paid by us or prohibition of the sale of our products and services, could result in a material adverse effect on our business, financial condition and results of operations.

      On June 15, 2005, Dr. Anthony W. Czarnik, a former employee, filed suit against us in the U.S. District Court for the District of Delaware seeking correction of inventorship of certain our patents and patent applications and alleging that we committed inequitable conduct and fraud in not naming him as an inventor. Dr. Czarnik seeks an order requiring us and the U.S. Patent and Trademark Office to correct the inventorship of certain of our patents and patent applications by adding Dr. Czarnik as an inventor, a judgment declaring certain of our patents and patent applications unenforceable, unspecified monetary damages and attorney’s fees. On August 4, 2005 we filed a motion to dismiss the complaint for lack of standing and failure to state a claim. While this motion was pending, Dr. Czarnik filed an amended complaint on September 23, 2005. On October 7, 2005, we filed a motion to dismiss the parties have pending motions before the court,amended complaint for lack of standing and failure to state a claim, and this motion is still pending. There has been no trial date has yet been set for this case. We believe we have meritorious defenses against this claim.

      No matters were submitted to a vote of security holders during the fourth quarter of 2004.2005.

      Our common stock has been quoted on the Nasdaq National Market under the symbol “ILMN” since July 28, 2000. Prior to that time, there was no public market for our common stock. The following table sets forth, for the periods indicated, the quarterly high and low sales prices per share of theour common stock as reported on the Nasdaq National Market. Our present policy is to retain earnings, if any, to finance future growth. We have never paid cash dividends and have no present intention to pay cash dividends in the foreseeable future.

      At January 31, 2005,2006, there were approximately 156229 stockholders of record, and the closing price per share of our common stock, as reported on the Nasdaq National Market on such date, was $9.69.$21.44.

      None.

      We currently have no plans or programs todid not repurchase sharesany of our stock. However, in September 2004, we repurchased shares of unvested stock in connection with the termination of an employee as set forth below.securities during 2005.

      On July 27, 2000, we commencedWe completed our initial public offering pursuant to a Registration Statement on Form S-1 (File No. 333-33922)of common stock in July 2000, resulting in net offering proceeds of $101.3 million. We will continue to use proceeds from our initial public offering to fund operations. Through January 2, 2005,1, 2006, we have used approximately $19.5$30.9 million to purchase property, plant and equipment, approximately $2.4 million for the acquisition of CyVera, and approximately $44.4$46.8 million to fund general operating expenses. The remaining balance is invested in a variety of interest-bearing instruments including U.S. Treasury securities, corporate debt securities and money market accounts.

      The following selected historical consolidated financial data havehas been derived from our audited consolidated financial statements. The balance sheet data as of January 1, 2006 and January 2, 2005 and December 28, 2003 and statementsstatement of operations data for each of the three years in the period ended January 2, 20051, 2006 are derived from audited consolidated financial statements included in this Annual Report on Form 10-K. The balance sheet data as of December 28, 2003, December 29, 2002, and December 30, 2001 and December 31, 2000 and statementsstatement of operations data for each of the two years in the period ended December 30, 200129, 2002 are derived from our audited consolidated financial statements that are not included in this report.Annual Report on Form 10-K. The Company’s fiscal year is 52 or 53 weeks ending the Sunday closest to December 31, with quarters of 13 or 14 weeks ending the Sunday closest to March 31, June 30, and September 30. The years ended January 1, 2006 and January 2, 2005 were 52 and 53 weeks, respectively. You should read this table in conjunction with Item 7, “Management’s Discussion and Analysis of Financial Condition and Results of Operations,” and Item 8, “Financial Statements and Supplementary Data.”

      See Note 1 of Notes to Consolidated Financial Statementsthe consolidated financial statements for an explanation of the determination of the number of shares used to compute basic and diluted net loss per share.

      The following discussion and analysis should be read with “Selected“Item 6. Selected Financial Data” and our consolidated financial statements and notes thereto included elsewhere in this Annual Report on Form 10-K. The discussion and analysis in this Annual Report on Form 10-K may contain contains forward-looking statements that involve risks and uncertainties, such as statements of our plans, objectives, expectations and intentions. Words such as “anticipate”, “believe,” “continue,” “estimate,” “expect,” “intend,” “may,” “plan,” “potential,” “predict,” “project” or similar words or phrases, or the negatives of these words, may identify forward-looking statements, but the absence of these words does not necessarily mean that a statement is not forward looking. Examples of forward-looking statements include, among others, statements regarding the integration of CyVera’s technology with our existing technology, the commercial launch of new products, including products based on CyVera’s technology, and the duration which our existing cash and other resources is expected to fund our operating activities. Forward-looking statements are subject to known and unknown risks and uncertainties and are based on potentially inaccurate assumptions that could cause actual results to differ materially from those expected or implied by the forward looking statements. Factors that could cause or contribute to these differences include those discussed in “Item 1A. Risk Factors” as well as those discussed elsewhere. The risk factors and other cautionary statements made in this Annual Report on Form 10-K should be read as applying to all related forward-looking statements wherever they appear in this Annual Report on Form 10-K. Our actual results could differ materially from those discussed here. Factors that could cause or contribute to these differences include those discussed in “Factors Affecting Operating Results” below as well as those discussed elsewhere.

      Illumina, Inc. wasWe were incorporated in April 1998. We develop and market next-generation tools for the large-scale analysis of genetic variation and function. Understanding genetic variation and function is critical to the development of personalized medicine, a key goal of genomics. Using our technologies, we have developed a comprehensive line of products that are designed to provide the performance, throughput, cost effectiveness and flexibility necessary to enable researchers in the life sciences and pharmaceutical industries to perform the billions of tests necessary to extract medically valuable information from advances in genomics. This information is expected to correlate genetic variation and gene function with particular disease states, enhancing drug discovery, allowing diseases to be detected earlier and more specifically, and permitting better choices of drugs for individual patients.

      In 2001, we began commercial sale of short pieces of DNA orcalled oligonucleotides, which we refer to as oligos, manufactured using our proprietary Oligator technology. We believe our Oligator technology is more cost effective than competing technologies, which has allowedand this advantage enabled us to market our oligonucleotidesoligos under a price leadership strategy while still achieving attractive gross margins.

      In 2001, we commercialized the first implementation of our BeadArray technology, the Sentrix Array Matrix. This is a disposable matrix with 96 fiber optic bundles arranged in a pattern that matches the standard96-well microtiter plate. Each fiber optic bundle performs more than 1,500 unique assays, which enables researchers to perform focused genotyping experiments in a high-throughput format. This format was also initiatedused to initiate our SNPsingle nucleotide polymorphism (“SNP”) genotyping services product line. As a result of the increasing market acceptance of our high throughput, low cost BeadArray technology, we have entered into genotyping services contracts with many leading genotyping centers, and have beenwere awarded $9.1 million from the National Institutes of Health to play a major role in the first phase of the International HapMap Project.

      Our production-scale BeadLab is baseda turnkey platform that includes all hardware and software necessary to enable researchers to perform genetic analysis research on the systemwhat we developed that has been operational in our genotyping service product line since 2001. In addition to our Sentrix Array Matrices, it includes the BeadArray Reader, a proprietary scanner that uses a laser to read the results of experiments captured on our arrays, as well as the GoldenGate SNP genotyping assay which can analyze up to 1536 SNPs per DNA sample.believe is an unprecedented scale. This system is being marketed to a small number of high throughputhigh-throughput genotyping users. As of January 2, 2005,1, 2006, we have installed and recorded revenue for nine11 BeadLabs.

      In 2003, we announced the launch of several new products, including 1) a new array format, the Sentrix BeadChip, which significantly expands market opportunities for our BeadArray technology and provides increased experimental flexibility for life science researchers; 2) a gene expression product line on both the Sentrix Array Matrix and the Sentrix BeadChip that allows researchers to analyze a focused set of genes across eight to 96 samples on a single array; and 3) a benchtop SNP genotyping and gene expression system, the BeadStation, for performing moderate-scale genotyping and gene expression using our technology. The BeadStation includes our BeadArray Reader, analysis software and assay reagents and is designed to match the throughput requirements and variable automation needs of individual research groups and core labs. Sales of these products began in the first quarter of 2004 and, as of January 2, 2005,1, 2006, we have shipped 42115 BeadStations.

      In late 2004, we announced a strategic collaboration with Invitrogen Corporation (“Invitrogen”) to synthesize and distribute oligos. In the launchthird quarter of new2005, we began shipping oligo products in connection with this agreement. As part of the agreement, we have developed the next generation of our Oligator DNA synthesis technology, which we have designed to support both plate- and tube-based capabilities. Invitrogen is responsible for sales, marketing and technical support. Profits from sales of collaboration products are divided equally between the two companies.

      In 2005, we began shipments of Sentrix BeadChips for whole-genome gene expression and whole-genome genotyping. The whole-genome gene expression BeadChips are designed to enable high-performance, cost-effective, whole-genome expression profiling of multiple samples on a single chip, resulting in a dramatic reduction in cost of whole-genome expression analysis while allowing researchers to expandanalysis. Our whole-genome expression product line includes multi-sample products for both the scaleHuman and reproducibility of large-scale biological experimentation.Mouse Genomes. The whole-genome genotyping BeadChip can be scaledis designed to unlimitedscale to high levels of multiplexing without compromising data quality and willto provide scientists the ability to query hundreds of thousands of SNPs in parallel. In 2004,the second quarter of 2005, we also announced two new versions of the Sentrix Array Matrix designed for researchers who want to take advantagecommenced shipment of our technology, but whose projects require fewerfirst whole-genome genotyping BeadChip, the HumanHap1, which interrogates more than 100,000 SNPs per sample than the number utilized on our standard 1536-plex array products.in parallel.

      In late 2004, we announced a strategic collaboration with Invitrogen Corporation to synthesize and distribute oligos. Under the agreement, we intend to expand our Oligator DNA synthesis technology to include both plate and tube based capability and Invitrogen will be responsible for sales, marketing and technical support. Profits from sales of collaboration products will be divided equally between the two companies.

      In January 2006, we began shipment of the new Sentrix HumanHap300 Genotyping BeadChip to customers around the world. Using the Infinium assay, which enables us to select virtually any SNP in the genome, the HumanHap300 BeadChip offers genomic coverage for more than 317,000 SNPs. We selected the SNP assays in collaboration with a consortium of scientists that are leaders in the genotyping field. We believe this product has quality and performance features that support our expectation that it will become an important discovery tool for researchers seeking to expand our customer base for our BeadArray technology; however, we can give no assurance that our sales efforts will continue to be successful.understand the genetic basis of common, yet complex diseases.

      Our revenues arerevenue is subject to fluctuations due to the timing of sales of high-value products and service projects, the impact of seasonal spending patterns, the timing and amount of government grant funding programs, the timing and size of research projects our customers perform, changes in overall spending levels in the life science industry, the timing and amount of government grant funding programs and other unpredictable factors that may affect our customer ordering patterns. Approximately 30% of our revenues for the year 2004 resulted from transactions that were funded under the International HapMap Project. We currently expect that most of the activities under this grant involving the Company and its customers will be completed in early 2005. We expect that the planned commercial launch of our whole genome genotyping and gene expression arrays, combined with the continued expansion of our existing product lines, will offset the loss of revenues funded by the HapMap grant and will drive future revenue growth. However, anyAny significant delays in the commercial launch or any lack or delay of thesecommercial acceptance of new products, unfavorable sales trends in our existing product lines, or impacts from the other factors mentioned above, could adversely affect our revenue growth in 20052006 or cause a sequential decline in quarterly revenues. Due to the possibility of fluctuations in our revenue and net income or loss, we believe quarterly comparisons of our operating results are not a good indication of our future performance.

      We have incurred substantial operating losses since our inception. As of January 2, 2005,1, 2006, our accumulated deficit was $123.7$144.6 million, and total stockholders’ equity was $72.3$72.5 million. These losses have principally occurred as a result of the substantial resources required for the research, development and manufacturing scale up effort required to commercialize our products and services, as well as chargesan acquired in-process research and development charge of $15.8 million related to our acquisition of CyVera and a charge of $5.9 million related to a termination-of-employmenttermination-of-employment lawsuit. We expect to continue to incur substantial costs for research, development and manufacturing scale up activities over the next several years. We will also need to significantly increase our selling, general and administrative costs as we build up our sales and marketing infrastructure to expand and support the sale of systems, other products and services. As a result of the expected increase in expenses, we will need to increase revenue significantly to achieve sustained profitability.

      Our discussion and analysis of our financial condition and results of operations is based upon our consolidated financial statements, which have been prepared in accordance with accounting principlesU.S. generally accepted in the United States.accounting principles. The preparation of financial statements requires that management make estimates, assumptions and judgments with respect to the application of accounting policies that affect the reported amounts of assets, liabilities, revenues and expenses, and the disclosures of contingent assets and liabilities. Actual results could differ from those estimates.

      Our significant accounting policies are described in Note 1 to our consolidated financial statements. Certain accounting policies are deemed critical if 1) they require an accounting estimate to be made based on assumptions that were highly uncertain at the time the estimate was made, and 2) changes in the estimate that are reasonably likely to occur, or different estimates that we reasonably could have used would have a material effect on our consolidated financial statements.

      Management has discussed the development and selection of these critical accounting estimatespolicies with the Audit Committee of our Board of Directors, and the Audit Committee has reviewed the disclosure. In addition, there are other items within our financial statements that require estimation, but are not deemed critical as defined above.

      We believe the following critical accounting policies reflect our more significant estimates and assumptions used in the preparation of the consolidated financial statements.

      Our sales arerevenue is generated primarily from two sources: product revenuethe sale of products and services revenue.services. Product revenue consists of sales of oligonucleotides, arrays, assay reagents, genotyping systemsinstrumentation, and gene expression systems. Servicesoligos. Service and other revenue consists of revenue received for performing genotyping services, and extended warranty sales.sales and revenue earned from milestone payments. As described below, significant judgments and estimates must be made and used in connection with the revenue recognized in any accounting period.

      We recognize revenue in accordance with the guidelines established by SEC Staff Accounting Bulletin (SAB)(“SAB”) No. 104. Under SAB No. 104, revenue cannot be recorded until all of the following criteria have been met: persuasive evidence of an arrangement exists; delivery has occurred or services have been rendered; the seller’s price to the buyer is fixed or determinable; and collectibility is reasonably assured.

      Product deliveryRevenue for product sales is recognized generally occurs when product is deliveredupon shipment and transfer of title to a common carrier or when the customer, receives the product, depending on the nature of the arrangement and provided no significant obligations remain.remain and collection of the receivables is reasonably assured. Revenue from the sale of instrumentation is recognized when earned, which is generally upon shipment. However, in the case of BeadLabs, are considered deliveredrevenue is recognized upon shipment,the completion of installation, training and fulfillmentthe receipt of contractually defined acceptance criteria and we need to determine the completion of each of these deliverables before revenue can be recognized. Genotypingcustomer acceptance. Revenue for genotyping services are considered deliveredis recognized when earned, which is generally at the time the genotyping analysis data is delivered to the customer. We have been awarded $9.1 million from the National Institutes of Health to perform genotyping services in connection with the first phase of the International HapMap Project. A portion of the services related to this project is considered delivered at the time the related costscustomer or as specific milestones are incurred while the remainder is considered delivered upon the delivery of genotyping data.achieved.

      In order to assess whether the price is fixed and determinable, we ensure there are no refund rights. If payment terms are based on future performance, we defer revenue recognition until the price becomes fixed and determinable. We assess collectibility based on a number of factors, including past transaction history with the customer and the creditworthiness of the customer. If we determine that collection of a payment is not reasonably assured, we defer revenue recognition until the time collection becomes reasonably assured, which is generally upon receipt of payment. Changes in judgments and estimates made in determining whether the criteria of SAB No. 104 have been met might result in a change in the timing or amount of revenue recognized.

      Sales of our genotyping and gene expression systemsinstrumentation generally include a standard one yearone-year warranty. We also sell separately priced maintenance (extended warranty) contracts, which are generally for one or two years, upon the expiration of the initial warranty. Revenue for extended warranty sales is recognized ratably over the term of the extended warranty.warranty period. Reserves are provided for estimated product warranty expenses at the time the associated revenue is recognized. If we were to experience an increase in warranty claims or if costs of servicing our warrantied products were greater than our estimates, our gross margins could be adversely affected.

      While the majority of our sales agreements contain standard terms and conditions, we do enter into agreements that contain multiple elements or non-standard terms and conditions. Emerging Issues Task Force (“EITF”) No. 00-21, (“EITF 00-21”), “RevenueRevenue Arrangements with Multiple Deliverables”,Deliverables, provides guidance on accounting for arrangements that involve the delivery or performance of multiple products, services, or rights to use assets within contractually binding arrangements. Significant contract interpretation is sometimes required to determine the appropriate accounting, including whether the deliverables specified in a multiple element arrangement should be treated as separate units of accounting for revenue recognition purposes, and if so, how the price should be allocated among the deliverable elements, when to recognize revenue for each element, and the period over which revenue should be recognized. We recognize revenue for delivered elements only when we believedetermine that the fair values of undelivered elements are known and there are no uncertainties regarding customer acceptance.

      A third source of revenue, research revenue, consists of amounts earned under research agreements with government grants, which is recognized in the period during which the related costs are incurred. All revenues arerevenue is recorded net of any applicable allowances for returns or discounts.

      We maintain an allowance for doubtful accounts for estimated losses resulting from the inability of our customers to make required payments. We evaluate the collectibility of our accounts receivable based on a combination of factors. We regularly analyze customer accounts, review the length of time receivables are outstanding and review historical loss rates. If the financial condition of our customers were to deteriorate, additional allowances could be required.

      We record adjustments to inventory for potentially excess, obsolete or impaired goods in order to state inventory at net realizable value. We must make assumptions about future demand, market conditions and the release of new products that will supercede old ones. We regularly review inventory for excess and obsolete products and components, taking into account product life cycle and development plans, product expiration and quality issues, historical experience and our current inventory levels. If actual market conditions are less favorable than anticipated, additional inventory adjustments could be required.

      We are subject to legal proceedings primarily related to intellectual property matters. Based on the information available at the balance sheet dates and through consultation with our legal counsel, we assess the likelihood of any adverse judgments or outcomes of these matters, as well as the potential ranges of probable losses. If losses are probable and reasonably estimable, we will record a reserveliability in accordance with Statement of Financial Accounting Standards (“SFAS”) No. 5, “AccountingAccounting for Contingencies”Contingencies. Currently, we have no such reservesliabilities recorded. Any reserves recordedThis may change in the future may change due todepending upon new developments in each matter.

      In December 2004, the Financial Accounting Standards Board (FASB)(“FASB”) issued FASB StatementSFAS No. 123 (revised 2004),Share Based Payment(“SFAS 123R)123R”), which is a revision of FASB StatementSFAS No. 123,Accounting for Stock-Based Compensation(SFAS 123). This statement supercedes APBAccounting Principles Bulletin (“APB”) Opinion No. 25,Accounting for Stock Issued to Employees,(APB 25), and amends FASB StatementSFAS No. 95,Statement of Cash Flows.Flows. Generally, the approach in SFAS No. 123R is similar to the approach described in SFAS No. 123; however, SFAS No. 123R requires all share-based payments to employees, including grants of employee stock options, to be recognized in the income statement based on their fair values. Pro forma disclosure is no longer an alternative.

      SFAS No. 123R permits companies to adopt its requirements using either a “modified prospective” method or a “modified retrospective” method. Under the “modified prospective” method, compensation cost is recognized in the financial statements beginning with the effective date, based on the requirements of SFAS No. 123R for all share-based payments granted after that date, and based on the requirements for SFAS No. 123 for all unvested awards granted prior to the effective date of SFAS No. 123R. Under the “modified retrospective” method, the requirements are the same as under the “modified prospective” method, but companies may restate financial statements of previous periods based on pro forma disclosures made in accordance with SFAS No. 123. We currently utilize the Black-Scholes model to measure the fair value of stock options granted to employees under the pro forma disclosure requirements of FASSFAS No. 123. While SFAS No. 123R permits companies to continue to use such model, it also permits the use of a “lattice” model. We have not yet determined which modeldeterimined we will use the Black-Scholes model to measure the fair value of employee stock options under the adoption for SFAS No. 123R. The new standard is effective for periodscompanies that are not small business issuers, like us, beginning with the first reporting period during the first fiscal year beginning on or after June 15, 2005, and we expect to adoptadopted SFAS No. 123R at the beginning of our new reporting period on July 4, 2005.January 2, 2006.

      We currently account for share-based payments to employees using APB No. 25’s intrinsic value method and, as such, recognize no compensation cost for employee stock options granted with exercise prices equal to or greater than the fair value of our common stock on the date of the grant. Accordingly, the adoption of SFAS No. 123R’s fair value method is expected to result in significant non-cash charges which will increase our reported operating expenses; however,expenses. However, it will have no impact on our cash flows. The precise impact of adoption of SFAS No. 123R cannot be predicted at this time because it will depend on the level of share-based payments granted in the future and the model we choose to use.future. However, had we adopted SFAS No. 123R in prior periods, we believe the impact of that standard would have approximated the impact of SFAS No. 123 as described in the disclosure of pro forma net loss under Stock-Based Compensation in Note 1the notes to our consolidated financial statements.

      To enhance comparability, the following table sets forth audited Consolidated Statementsconsolidated statement of Operationsoperations data for the years ended January 1, 2006, January 2, 2005, and December 28, 2003 and December 29, 2002 stated as a percentage of total revenue.

      We recorded a $7.7 million charge in June 2002 to cover total damages and estimated expenses related to a jury verdict in atermination-of-employment lawsuit. We appealed the decision, and in December 2004, the Fourth Appellate District Court of Appeal, in San Diego, California, reduced the amount of the award. During the appeal process, the court required us to incur interest charges on the judgment amount at statutory rates until the case was resolved. During the years ended January 2, 2005 and December 28, 2003, we recorded $0.6 million and $0.8 million, respectively, of such interest charges as litigation expense. As a result of the revised judgment, we reduced the $9.2 million liability on our balance sheet to $5.9 million and recorded a gain of $3.3 million as a litigation judgment in the fourth quarter of 2004. In addition, in August 2004, we recorded a $1.5 million gain as a result of a settlement with Applera.

      Our fiscal year is 52 or 53 weeks ending the Sunday closest to December 31, with quarters of 13 or 14 weeks ending the Sunday closest to March 31, June 30, and September 30. The years ended January 2, 2005 and December 28, 2003 arewere 53 and 52 weeks, respectively.

      Revenue for the years ended January 2, 2005 and December 28, 2003 was $50.6 million and $28.0 million, respectively. Product revenue increased to $40.5 million in 2004 from $18.4 million in 2003. The increase resulted almost entirely from sales of consumables used on our BeadLabs and BeadStations and sales of our benchtop BeadStations, offset by fewer sales of our production-scale BeadLabs. In 2003, we had no sales of BeadStations and we only began selling consumable products in May 2003.

      Service revenue increased to $8.1 million in 2004for the year ended January 2, 2005 from $6.5 million in for the year ended December 28, 2003. Substantially all of this increase relates to SNP genotyping services performed for the International HapMap Project. We are the recipient of a grant from the National Institutes of Health covering our participation in the first phase of the International HapMap Project, which is a $100 million internationally funded successor project to the Human Genome Project that will help identify a map of genetic variations that may be used to perform disease-related research. We could receive up toreceived $9.1 million of funding for this project which coverscovered basic research activities, the development of SNP assays and the genotyping to be performed on those assays. We havehad recognized revenue underfrom this grant of $8.4$8.3 million and, as ofthrough the end of 2004, we had approximately $0.72004. The remaining $0.8 million of funding remaining related to this project which is expected to bewas received and recognized as revenue in early 2005.

      Government grants and other research funding decreased to $2.0 million for the year ended January 2, 2005 from $3.2 million for the year ended December 28, 2003, primarily due to a decrease in internal research spending for our grant from the National Institutes of Health covering our participation in the International HapMap Project. We expect government grants to decline as a percentage of total revenues.